CN102625109A - Multi-core-processor-based moving picture experts group (MPEG)-2-H.264 transcoding method - Google Patents
Multi-core-processor-based moving picture experts group (MPEG)-2-H.264 transcoding method Download PDFInfo
- Publication number
- CN102625109A CN102625109A CN2012100894359A CN201210089435A CN102625109A CN 102625109 A CN102625109 A CN 102625109A CN 2012100894359 A CN2012100894359 A CN 2012100894359A CN 201210089435 A CN201210089435 A CN 201210089435A CN 102625109 A CN102625109 A CN 102625109A
- Authority
- CN
- China
- Prior art keywords
- macro block
- block group
- responsible
- information
- mpeg
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The invention discloses a multi-core-processor-based moving picture experts group (MPEG)-2-H.264 transcoding method, which comprises the following steps of: (1) receiving a code stream by using a Core 1, performing entropy decoding by using a Core 2, and performing macro block decoding by using a Core 3; (2) performing macro block estimation by using Cores 4 to 9, and performing macro block prediction by using Cores 10 to 12; (2) performing macro block transformation quantification and inverse quantification inverse transformation by using the Cores 11 to 12, and performing macro block filtering by using Cores 15 to 16; and (4) performing entropy encoding by using a Core 13, and transmitting the code stream by using a Core 14. According to the transcoding method, tasks are allocated to each core by adopting a function and data combination division mode according to relationships between each module of a cascaded pixel domain transcoding (CPDT) structure, so that the parallel speed-up ratio of a transcoding process can be effectively increased; and the method is suitable for high-code rate MPEG-2-H.264 transcoding.
Description
Technical field
The invention belongs to the Video Transcoding Technology field, be specifically related to a kind of MPEG-2 that realizes based on polycaryon processor code-transferring method extremely H.264.
Background technology
Along with the popularization of digital TV in high resolution (HDTV), the development of HD video treatment technology is more and more rapider.The vision signal of tradition HDTV mainly adopts MPEG-2 to compress, and shortcoming is that compression ratio is not high, in network transmission process, can consume a large amount of bandwidth.H.264 as video compression standard of new generation, have excellent compression performance, in HD video process field in recent years, obtained extensive use.Therefore, MPEG-2 HD video transcoding extremely H.264 has great using value; But because the high definition transcoding amount of calculation is huge; Common single core processor is difficult to real-time implementation; The MPEG-2 high-definition real-time transcoding extremely H.264 that develops into of polycaryon processor provides the foundation, but the parallel restraining factors that become key of how to divide the process information of monokaryon and how to realize multinuclear efficiently.
The HD video trans-coding system is divided into the decoded portion of MPEG-2 and coded portion H.264.Therefore characteristics such as the structure of transcoding device can be divided into cascade pixel domain code conversion structure (CPDT) and discrete cosine transform domain transcoding structure (DDT), and with respect to the DDT structure, it is flexible that CPDT has structure, and drift error is little are widely used in the HD video trans-coding system.
Traditional CPDT transcoding structure is separated with decoded portion and coded portion usually, and in the multi-core parallel concurrent of various piece is realized, adopts based on single data division or the method for dividing based on single function.
Single data divide to realize normally being that unit carries out with the macro block, in the decoded portion of MPEG2, need at first carry out the entropy decoding, just can carry out data then and divide.At coded portion H.264, need frame of video is divided into a plurality of slice (sheet) slice of each nuclear coding.Such division, the hunting zone when having reduced video coding, loss to some extent on search accuracy and compression efficiency, and changed the structure of encoding stream.And, increased the load of nuclear because the division of slice makes that the part filtering needs after the frame of video coding is accomplished, to carry out again.In sum, single data are divided, parallel weak effect, and the transcoding effect also can be lost.
It is that unit carries out with the encoding and decoding basic function module normally that single function is divided, and its basic thought is on single nuclear, to realize a transcoding function, and transcoding begins to form flowing water behind the certain hour, thereby realizes parallel.In the CPDT trans-coding system, MPEG-2 decoding basic module comprises entropy decoding (VLD), inverse discrete cosine transform (IDCT), inverse quantization (IQ) and motion compensation (MC_D); H.264 the basic module of encoding comprises that estimation (ME), intra prediction mode are estimated (IPE), motion compensation (MC_E), infra-frame prediction (IP), discrete cosine transform (DCT) and inverse discrete cosine transform (IDCT), quantification (Q) and inverse quantization (IQ), entropy coding (EC), Rate Control (RC), filtering strength (BS) calculates and the interior block elimination filtering (LPF) of ring.Because the complexity of each functional module differs, operand differs greatly, if the form that only adopts simple function to divide, each nuclear is accomplished a function, in the time of must causing function synchronous, waits for each other for a long time between nuclear and the nuclear.The inevitable too early state that is in wait of nuclear that operand is less is unfavorable for the raising of parallel efficiency.
Summary of the invention
To the above-mentioned technological deficiency of existing in prior technology, the invention provides a kind of MPEG-2 that realizes based on polycaryon processor code-transferring method extremely H.264, can effectively improve the parallel speed-up ratio of transcoding process.
A kind of MPEG-2 that realizes based on polycaryon processor code-transferring method extremely H.264, described polycaryon processor has 16 Core (kernel) at least, each Core parallel processing processor active task separately; This method comprises the steps:
(1) be responsible for receiving the code stream of MPEG-2 video format by Core1, be responsible for described code stream is carried out the entropy decoding and generates the several macro blocks group successively by Core2, described macro block group is made up of n continuous macro block, and n is the natural number greater than 0;
Obtain the decoded data of macro block group by responsible described macro block group is decoded of Core3;
(2) be responsible for described decoded data is estimated to obtain the estimated information of macro block group by Core4～9, described estimated information is predictive vector or intra prediction mode;
Be responsible for predicting the information of forecasting that obtains the macro block group by Core10～12, make described decoded data deduct the residual information that information of forecasting obtains the macro block group according to described estimated information;
(3) be responsible for described residual information is carried out transform and quantization successively by Core11～12; Be responsible for the residual information after quantizing rebuild successively by Core11～12 and Core15～16 and obtain the reconstructed image of macro block group with filtering and store;
(4) by Core13 be responsible for to described estimated information with quantize after residual information carry out entropy coding and obtain the H.264 code stream of video format of macro block group corresponding a section, and be responsible for sending this segment encode stream by Core14.
Preferably, described macro block group is made up of 16 continuous macro blocks; The calculation process performance of each Core is at utmost embodied.
The code stream of MPEG-2 video format includes residual error data and motion vector through the macro block that the entropy decoding generates; In the described step (1), the process that the macro block group is decoded is: at first the motion vector according to the macro block group carries out the prediction data that motion compensation obtains the macro block group; Then the residual error data of macro block group is carried out inverse quantization and inverse transformation successively; Make residual error data and prediction data addition after the macro block group inverse transformation at last, obtain the decoded data of macro block group.
In the described step (2), decoded data is estimated to be divided into two kinds of situation:
If the image under the macro block group is a key frame, then is responsible for that by Core8～9 decoded data is carried out intra prediction mode and estimates to obtain intra prediction mode;
If the image under the macro block group is non-key frame, then is responsible for decoded data is carried out estimation and intra prediction mode estimation, and after relatively, obtains predictive vector or intra prediction mode by Core4～9.
Preferably, the motion vector according to the macro block group carries out estimation to decoded data; Can eliminate the dependence of confirming the motion search starting point between the adjacent macroblocks, the amount of calculation that reduces estimation improves corresponding speed.
Preferably; If the image under the macro block group is non-key frame; Then be responsible for respectively the decoded data of n/4 macro block in the macro block group is carried out estimation, be responsible for respectively the decoded data of n/2 macro block in the macro block group is carried out the intra prediction mode estimation by each Core in Core8～9 simultaneously by each Core in Core4～7; If the image under the macro block group is a key frame, then be responsible for respectively the decoded data of n/2 macro block in the macro block group is carried out the intra prediction mode estimation by each Core in Core8～9.Computational load that can balance Core4～9 is optimized whole transcoding performance.
In the described step (2), predict the two kinds of situation that are divided into according to estimated information:
If the estimated information of macro block group is an intra prediction mode, then be responsible for carrying out infra-frame prediction according to intra prediction mode by Core11～12, obtain the information of forecasting of macro block group;
If the estimated information of macro block group is a predictive vector, then be responsible for carrying out motion compensation according to predictive vector by Core10, obtain the information of forecasting of macro block group.
In the described step (3), rebuild with the process of filtering following successively to the residual information after quantizing:
A. be responsible for the residual information after quantizing is carried out inverse quantization and inverse transformation successively by Core11～12;
B. make residual information and the information of forecasting addition of macro block group after the inverse transformation, obtain the reconstruction information of macro block group; Be responsible for that by Core15 described reconstruction information is carried out filtering strength and calculate filter strength value;
C. be responsible for reconstruction information being encircled interior block elimination filtering by Core16, obtain the reconstructed image of macro block group according to described filter strength value.
Preferably, Core11 and Core12 are responsible for the luminance component and the corresponding calculation process task of chromatic component of each macro block in the macro block group respectively; Computational load that can balance Core11～12 is optimized whole transcoding performance.
Be responsible for the quantizing process in the step (3) being carried out Rate Control by Core14 according to the bit rate output of entropy coding process in the step (4).
Describedly be transformed to discrete cosine transform, contravariant is changed to inverse discrete cosine transform.
Code-transferring method of the present invention is according to the correlation of each intermodule of CPDT transcoding structure; The dividing mode that adopts function and data to combine is carried out Task Distribution to each nuclear; Can effectively improve the parallel speed-up ratio of transcoding process, and be more suitable for transcoding extremely H.264 in the MPEG-2 of high code check.
Description of drawings
Fig. 1 is the execution schematic flow sheet of code-transferring method of the present invention.
Fig. 2 is the structural representation of CPDT transcoding device.
Fig. 3 divides the processing time sketch map of each module afterwards for CPDT transcoding device combines based on function and data.
Embodiment
In order to describe the present invention more particularly, coding/decoding method of the present invention is elaborated below in conjunction with accompanying drawing and embodiment.
As shown in Figure 1, a kind of MPEG-2 that realizes based on polycaryon processor code-transferring method extremely H.264, polycaryon processor adopts the polycaryon processor of Tilera company's T ilePro64 series, and processor has 64 Core, each Core parallel processing processor active task separately; This method comprises the steps:
(1) be responsible for receiving the code stream of MPEG-2 video format by Core1, be responsible for code stream is carried out entropy decoding (VLD) and generates the several macro blocks group successively by Core2, the macro block group is made up of 16 continuous macro blocks.
The code stream of MPEG-2 video format includes residual error data and motion vector through the macro block that the entropy decoding generates; Obtain the decoded data of macro block group by the responsible macro block group is decoded of Core3: at first the motion vector according to the macro block group carries out the prediction data that motion compensation (MC_D) obtains the macro block group; Then the residual error data of macro block group is carried out inverse quantization and inverse discrete cosine transform successively; Make residual error data and prediction data addition after the macro block group inverse transformation at last, obtain the decoded data of macro block group.
When current macro was carried out motion compensation, the decoded data that need get reference frame as a reference; The purpose of inverse quantization (IQ) and inverse discrete cosine transform (IDCT) is to remove the spatial redundancy and control code check of original image, and amount of compressed data is as far as possible intactly replied original image.
(2) by the responsible estimated information (predictive vector or intra prediction mode) of the decoded data of macro block group being estimated to obtain the macro block group in Core4～9:
If the image under the macro block group is non-key frame; Then be responsible for respectively the decoded data of 4 macro blocks in the macro block group is carried out estimation according to the motion vector of macro block group by each Core in Core4～7; Be responsible for respectively the decoded data of 8 macro blocks in the macro block group is carried out the intra prediction mode estimation by each Core in Core8～9 simultaneously; Obtain the predictive vector and the intra prediction mode of macro block group, after relatively, choose one of which;
If the image under the macro block group is a key frame, then be responsible for respectively the decoded data of 8 macro blocks in the macro block group is carried out the intra prediction mode estimation by each Core in Core8～9;
Estimation (ME) is used in inter prediction, in reference frame, searches for the position of the reference macroblock that matees the most from current coding macro block, and displacement is exactly a predictive vector between the macro block; Intra prediction mode estimates that (IPE) selects suitable intra prediction mode in order to carry out next step infra-frame prediction.
Be responsible for predicting the information of forecasting that obtains the macro block group by Core10～12 according to the estimated information of macro block group:
If the estimated information of macro block group is an intra prediction mode, then be responsible for carrying out infra-frame prediction (IP) according to the intra prediction mode of macro block group by Core11～12, obtain the information of forecasting of macro block group; Infra-frame prediction is according to different patterns, uses the data of the current macro left side one row and top delegation, and current coded macroblocks is predicted;
If the estimated information of macro block group is a predictive vector, then be responsible for carrying out motion compensation (MC_E) according to the predictive vector of macro block group by Core10, obtain the information of forecasting of macro block group; When current macro was carried out motion compensation, the reconstructed image that need get reference frame as a reference.
Make the decoded data of macro block group deduct the residual information that information of forecasting obtains the macro block group.
(3) be responsible for the residual information of macro block group is carried out discrete cosine transform and quantification successively by Core11～12.
Discrete cosine transform (DCT) is that time-domain information is mapped on the frequency domain, and to remove the correlation of signal, DCT and IDCT are a pair of reciprocal processes; Quantizing (Q) is the dynamic range that under the prerequisite that does not influence visual effect, reduces the frame of video coding, thereby reduces code check, improves compression ratio, and quantization principles H.264 is following:
Wherein: y is an image input sample, and FQ is the quantized value of y, and QP is a quantization step, and round is a bracket function.IQ is exactly the anti-process of Q: Y=FQ * QP; The process that quantizes is irreversible, and result that inverse quantization obtains and initial data are discrepant.
Be responsible for the residual information after quantizing is carried out inverse quantization and inverse discrete cosine transform successively by Core11～12; Make residual information and the information of forecasting addition of macro block group after the inverse transformation, obtain the reconstruction information of macro block group.
With the macro block is after unit carries out data compression and recovers, can produce the discontinuous phenomenon of image in macroblock boundaries, and this phenomenon is called blocking artifact; So need according to macro block (mb) type, the filter strength value on computing macro block border, the filtering strength according to each bar border carries out block elimination filtering with filter to reconstruction information then, eliminates blocking artifact.
Calculate filter strength value so carry out filtering strength (BS) by the responsible reconstruction information of Core15 to the macro block group; Be responsible for reconstruction information being encircled interior block elimination filtering (LPF) by Core16, obtain the reconstructed image of macro block group according to filter strength value; Reconstructed image is stored, used when frame supplies in the follow-up H.264 cataloged procedure estimation and motion compensation as a reference.
In this execution mode, Core11 and Core12 are responsible for the luminance component and the corresponding calculation process task of chromatic component of each macro block in the macro block group respectively.
(4) by Core13 be responsible for to the estimated information of macro block group with quantize after residual information carry out entropy coding (EC) and obtain the H.264 code stream of video format of macro block group corresponding a section, and be responsible for sending this segment encode stream by Core14.Entropy coding is the coding method of lossless compress, can undistorted decoding obtain original video through the code stream of entropy coding.
Be responsible for the quantizing process in the step (3) being carried out Rate Control (RC) by Core14 according to the bit rate output of entropy coding process.
The modular structure of CPDT trans-coding system is as shown in Figure 2, and whole system can be divided into MPEG-2 decoded portion and coded portion H.264.H.264, during this execution mode will be encoded these several couplings of DCT, Q, IQ and IDCT closely module can be combined into a computing module (CALC).Consider the data after the IP module need be used the IDCT reconstruction, therefore it also is incorporated in the CALC module.From the angle of balancing operational load, this execution mode is merged into a functional module with MPEG-2 decoded portion operand less I Q, IDCT and MC_D, and RC module that operand is few and code stream transmission task merge.
This execution mode is also introduced the notion that deal with data is divided; ME and the IPE module big to operand further split; IPE is split into IPE0 and IPE1; Each submodule is handled 8 macro blocks respectively, and two sub-module are parallel carries out IPE, and wherein the used macro block of the IPE left side one row pixel and top one-row pixels all adopt the respective pixel point of present frame; ME is split as ME0, ME1, ME2, ME3, and each submodule is handled 4 macro blocks respectively.Dependence when removing that the motion search starting point is confirmed between the adjacent macroblocks, the motion vector that decodes with MPEG-2 is as the search center of ME, the parallel ME that carries out of four sub-module.The CALC module of coded portion is split as CALC_L that handles brightness data and the CALC_C that handles chroma data; The processing time situation of each functional module is as shown in Figure 3 after dividing through deal with data, and the operand between visible each module has reached good balance, and the distribution of the corresponding nuclear of each functional module difference is as shown in table 1.
Table 1
| Functional module | The Core numbering |
| Receive |
1 |
| |
2 |
| |
3 |
| ME0、ME1、ME2、 |
4、5、6、7 |
| IPE0、 |
8、9 |
| MC_E | 10 |
| CALC_L、CALC_C | 11、12 |
| EC | 13 |
| RC& sends |
14 |
| BS | 15 |
| LPF | 16 |
Through experiment test; At code check is 8M; Under the situation that 16 caryogamy are put, adopt the existing multi-core parallel concurrent transcoding technology of dividing based on simple function with this execution mode same MPEG-2 code stream to be carried out transcoding, the parallel speed-up ratio of prior art is 11; And the parallel speed-up ratio of this execution mode has reached 12.4, compares and has improved 13%.
Claims (10)
1. a MPEG-2 who realizes based on polycaryon processor is to H.264 code-transferring method, and described polycaryon processor has 16 Core at least, each Core parallel processing processor active task separately; This method comprises the steps:
(1) be responsible for receiving the code stream of MPEG-2 video format by Core1, be responsible for described code stream is carried out the entropy decoding and generates the several macro blocks group successively by Core2, described macro block group is made up of n continuous macro block, and n is the natural number greater than 0;
Obtain the decoded data of macro block group by responsible described macro block group is decoded of Core3;
(2) be responsible for described decoded data is estimated to obtain the estimated information of macro block group by Core4～9, described estimated information is predictive vector or intra prediction mode;
Be responsible for predicting the information of forecasting that obtains the macro block group by Core10～12, make described decoded data deduct the residual information that information of forecasting obtains the macro block group according to described estimated information;
(3) be responsible for described residual information is carried out transform and quantization successively by Core11～12; Be responsible for the residual information after quantizing rebuild successively by Core11～12 and Core15～16 and obtain the reconstructed image of macro block group with filtering and store;
(4) by Core13 be responsible for to described estimated information with quantize after residual information carry out entropy coding and obtain the H.264 code stream of video format of macro block group corresponding a section, and be responsible for sending this segment encode stream by Core14.
2. the MPEG-2 that realizes based on polycaryon processor according to claim 1 code-transferring method extremely H.264, it is characterized in that: described macro block group is made up of 16 continuous macro blocks.
3. the MPEG-2 that realizes based on polycaryon processor according to claim 1 code-transferring method extremely H.264; It is characterized in that: in the described step (1), the process that the macro block group is decoded is: at first the motion vector according to the macro block group carries out the prediction data that motion compensation obtains the macro block group; Then the residual error data of macro block group is carried out inverse quantization and inverse transformation successively; Make residual error data and prediction data addition after the macro block group inverse transformation at last, obtain the decoded data of macro block group.
4. the MPEG-2 that realizes based on polycaryon processor according to claim 1 code-transferring method extremely H.264 is characterized in that: in the described step (2), decoded data is estimated to be divided into two kinds of situation:
If the image under the macro block group is a key frame, then is responsible for that by Core8～9 decoded data is carried out intra prediction mode and estimates to obtain intra prediction mode;
If the image under the macro block group is non-key frame, then is responsible for decoded data is carried out estimation and intra prediction mode estimation, and after relatively, obtains predictive vector or intra prediction mode by Core4～9.
5. the MPEG-2 that realizes based on polycaryon processor according to claim 4 code-transferring method extremely H.264, it is characterized in that: the motion vector according to the macro block group carries out estimation to decoded data.
6. the MPEG-2 that realizes based on polycaryon processor according to claim 4 code-transferring method extremely H.264; It is characterized in that: if the image under the macro block group is non-key frame; Then be responsible for respectively the decoded data of n/4 macro block in the macro block group is carried out estimation, be responsible for respectively the decoded data of n/2 macro block in the macro block group is carried out the intra prediction mode estimation by each Core in Core8～9 simultaneously by each Core in Core4～7; If the image under the macro block group is a key frame, then be responsible for respectively the decoded data of n/2 macro block in the macro block group is carried out the intra prediction mode estimation by each Core in Core8～9.
7. the MPEG-2 that realizes based on polycaryon processor according to claim 1 code-transferring method extremely H.264 is characterized in that: in the described step (2), predict the two kinds of situation that are divided into according to estimated information:
If the estimated information of macro block group is an intra prediction mode, then be responsible for carrying out infra-frame prediction according to intra prediction mode by Core11～12, obtain the information of forecasting of macro block group;
If the estimated information of macro block group is a predictive vector, then be responsible for carrying out motion compensation according to predictive vector by Core10, obtain the information of forecasting of macro block group.
8. the MPEG-2 that realizes based on polycaryon processor according to claim 1 code-transferring method extremely H.264 is characterized in that: in the described step (3), rebuild with the process of filtering following successively to the residual information after quantizing:
A. be responsible for the residual information after quantizing is carried out inverse quantization and inverse transformation successively by Core11～12;
B. make residual information and the information of forecasting addition of macro block group after the inverse transformation, obtain the reconstruction information of macro block group; Be responsible for that by Core15 described reconstruction information is carried out filtering strength and calculate filter strength value;
C. be responsible for reconstruction information being encircled interior block elimination filtering by Core16, obtain the reconstructed image of macro block group according to described filter strength value.
9. according to claim 1, the 7 or 8 described MPEG-2 that realize based on polycaryon processor code-transferring method extremely H.264, it is characterized in that: Core11 and Core12 are responsible for the luminance component and the corresponding calculation process task of chromatic component of each macro block in the macro block group respectively.
10. the MPEG-2 that realizes based on polycaryon processor according to claim 1 code-transferring method extremely H.264 is characterized in that: be responsible for according to the bit rate output of entropy coding process in the step (4) quantizing process in the step (3) being carried out Rate Control by Core14.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210089435.9A CN102625109B (en) | 2012-03-30 | 2012-03-30 | Multi-core-processor-based moving picture experts group (MPEG)-2-H.264 transcoding method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210089435.9A CN102625109B (en) | 2012-03-30 | 2012-03-30 | Multi-core-processor-based moving picture experts group (MPEG)-2-H.264 transcoding method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102625109A true CN102625109A (en) | 2012-08-01 |
| CN102625109B CN102625109B (en) | 2014-04-16 |
Family
ID=46564770
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210089435.9A Active CN102625109B (en) | 2012-03-30 | 2012-03-30 | Multi-core-processor-based moving picture experts group (MPEG)-2-H.264 transcoding method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102625109B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103796063A (en) * | 2012-10-30 | 2014-05-14 | 美国博通公司 | Parallel transcoding |
| CN106358043A (en) * | 2015-07-15 | 2017-01-25 | 联发科技股份有限公司 | Hybrid video decoding apparatus and associated hybrid video decoding method |
| CN106375767A (en) * | 2015-07-24 | 2017-02-01 | 联发科技股份有限公司 | Hybrid video decoder and associated hybrid video decoding method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080310515A1 (en) * | 2007-06-14 | 2008-12-18 | Yasutomo Matsuba | MPEG-2 2-Slice Coding for Simple Implementation of H.264 MBAFF Transcoder |
| CN102325255A (en) * | 2011-09-09 | 2012-01-18 | 深圳市融创天下科技股份有限公司 | Multi-core CPU (central processing unit) video transcoding scheduling method and multi-core CPU video transcoding scheduling system |
-
2012
- 2012-03-30 CN CN201210089435.9A patent/CN102625109B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080310515A1 (en) * | 2007-06-14 | 2008-12-18 | Yasutomo Matsuba | MPEG-2 2-Slice Coding for Simple Implementation of H.264 MBAFF Transcoder |
| CN102325255A (en) * | 2011-09-09 | 2012-01-18 | 深圳市融创天下科技股份有限公司 | Multi-core CPU (central processing unit) video transcoding scheduling method and multi-core CPU video transcoding scheduling system |
Non-Patent Citations (2)
| Title |
|---|
| LIU YU, LU WEI, WANG LI, ET AL.: "Design of MPEG-2 to H.264/AVC Transcoder", 《PROC. OF THE 10TH INTERNATIONAL SYMPOSIUM ON CONSUMER ELECTRONICS》 * |
| 杜耀刚, 蔡安妮: "DCT域视频转码技术综述", 《电子学报》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103796063A (en) * | 2012-10-30 | 2014-05-14 | 美国博通公司 | Parallel transcoding |
| CN103796063B (en) * | 2012-10-30 | 2017-04-12 | 安华高科技通用Ip(新加坡)公司 | Parallel transcoding |
| CN106358043A (en) * | 2015-07-15 | 2017-01-25 | 联发科技股份有限公司 | Hybrid video decoding apparatus and associated hybrid video decoding method |
| CN106375767A (en) * | 2015-07-24 | 2017-02-01 | 联发科技股份有限公司 | Hybrid video decoder and associated hybrid video decoding method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102625109B (en) | 2014-04-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP2023179682A (en) | Intra prediction using linear or affine transformation with adjacent sample reduction | |
| CN104221377B (en) | A kind of video decoding method, device and computer-readable storage medium | |
| CN103190147B (en) | For combined decoding method and the equipment of the syntactic element of video coding | |
| CN101159875B (en) | Double forecast video coding/decoding method and apparatus | |
| US20100172409A1 (en) | Low-complexity transforms for data compression and decompression | |
| CN104041045A (en) | Secondary boundary filtering for video coding | |
| US20140010284A1 (en) | Image transform and inverse transform method, and image encoding and decoding device using same | |
| CN105379270A (en) | Inter-color component residual prediction | |
| CN108028931A (en) | Method and device for the adaptive inter prediction of coding and decoding video | |
| CN101490968A (en) | Parallel processing apparatus for video compression | |
| CN104737542A (en) | Content adaptive entropy coding for next generation video | |
| CN103621082A (en) | Quantization in video coding | |
| CN101911702A (en) | Quantization based on rate-distortion modeling for cabac coders | |
| US8594189B1 (en) | Apparatus and method for coding video using consistent regions and resolution scaling | |
| CN103918263A (en) | Device and methods for scanning rectangular-shaped transforms in video coding | |
| CN105474642A (en) | Re-encoding image sets using frequency-domain differences | |
| WO2023040600A1 (en) | Image encoding method and apparatus, image decoding method and apparatus, electronic device, and medium | |
| CN101663895B (en) | Video coding mode selection using estimated coding costs | |
| CN106688234B (en) | Processor system, Video Codec and method for image transformation | |
| KR20100079037A (en) | Method and apparatus for intra encoding by fast intra mode searching | |
| CN103347185A (en) | Unmanned aerial vehicle scouting image synthesis compressed encoding method based on selective block transformation | |
| CN104704826A (en) | Two-step quantization and coding method and apparatus | |
| US10382758B2 (en) | Transform block-level scan order selection for video coding | |
| CN104937934A (en) | Autoregressive pixel prediction in the neighborhood of image borders | |
| KR20110123699A (en) | Method for coding a picture sequence, corresponding method for reconstruction and stream of coded data representative of said sequence |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |